U.S. patent number 5,911,983 [Application Number 08/466,597] was granted by the patent office on 1999-06-15 for gene therapy for gaucher disease using retroviral vectors.
This patent grant is currently assigned to University of Pittsburgh. Invention is credited to Alfred B. Bahnson, John A. Barranger, Paul Robbins.
United States Patent |
5,911,983 |
Barranger , et al. |
June 15, 1999 |
Gene therapy for Gaucher disease using retroviral vectors
Abstract
The present invention relates to gene therapy for Gaucher
disease using retroviral vectors which express the
glucocerebrosidase gene. Methods are provided for transduction of
autologous hematopoietic stem cells (e.g., human CD34+ cells) with
these vectors and for transplantation of the transduced cells into
a Gaucher disease patient to provide therapeutically effective
levels of glucocerebrosidase activity. The invention also provides
for retroviral vectors that express the glucocerebrosidase gene,
and for human hematopoietic cells that contain the retroviral
vector.
Inventors: |
Barranger; John A. (Gibsonia,
PA), Robbins; Paul (Pittsburgh, PA), Bahnson; Alfred
B. (Pittsburgh, PA) |
Assignee: |
University of Pittsburgh
(Pittsburgh, PA)
|
Family
ID: |
25419824 |
Appl.
No.: |
08/466,597 |
Filed: |
June 6, 1995 |
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
904809 |
Jun 26, 1992 |
|
|
|
|
Current U.S.
Class: |
424/93.21;
435/320.1; 435/372; 424/93.6 |
Current CPC
Class: |
C12Y
302/01045 (20130101); C12N 9/2402 (20130101); A61K
48/00 (20130101) |
Current International
Class: |
C12N
9/24 (20060101); A61K 48/00 (20060101); A61K
048/00 (); C12N 005/10 (); C12N 015/86 () |
Field of
Search: |
;435/69.1,172.1,172.3,320.1,240.2,366,372,372.2,372.3
;424/93.6,93.2,93.21 |
Foreign Patent Documents
Other References
"Report and Recommendations of the Panel to Assess the NIH
Investment in Gene Therapy", S.H. Orkin and A.G. Arno, Eds, Dec.
1995. .
Patrick, A.D., Biochem. J. 97:17C-18C (1965). .
Brady, R.O., et al., Biochem. Biophys. Res. Commun. 18:221-225
(1965). .
Erickson, A.H., et al., J. Biol. Chem. 260:14319-14324 (1985).
.
Ho, M.W., et al., Proc. Natl. Acad. Sci. USA 68:2810-2813 (1971).
.
O'Brien, J.S., et al., Science 241:1098-1101 (1988). .
Ginns, E.I., et al., Biochem. Biophys. Res. Commun. 123:574-580
(1984). .
Sorge, J., et al., Proc. Nat. Acad. Sci. USA 82:7289-7293 (1985).
.
Tsuji, S., et al., J. Biol. Chem. 261:50-53 (1986). .
Ginns, E.I., et al., Proc. Nat. Acad. Sci. USA 82:7101-7105 (1985).
.
Tsuji, S., et al., New Engl. J. Med. 316:570-575 (1987). .
Tsuji, S., et al., Proc. Natl. Acad. Sci. USA 85:2349-2352 (1988).
.
Firon, N., et al., Am. J. Hum. Genet. 46:527-532 (1990). .
Rappeport, J.M., et al., Birth Defects: Original Article Series
22,1:101-109 (1986). .
Barranger, J.A., et al., Japanese J. of Inher. Met. Disease
51:45-71 (1989). .
Takasaki, S., et al., J. Biol. Chem. 259:10112-10117 (1984). .
Furbish, F.S., et al., Biochem. Biophys. Acta. 673:425-434 (1981).
.
Miller, A.D., Blood 76:271-278 (1990). .
Miller, D.G., et al., Mol. Coll. Biol. 10:4239-4242 (1990). .
Bodine, D.M., et al., Exp. Hematol. 19:206-212 (1991). .
Nolta, J.A., et al., Blood 75:787-797 (1990). .
Correll, P.H., et al., Proc. Natl. Acad. Sci. USA 86:8912-8916
(1989). .
Scherdin, U., et al., J. Virol. 64:907-912 (1990). .
McLaughlin, S.K., et al., J. Virol. 65:1963-1973 (1988). .
Hunter, L.A., et al., J. Virol. 66:317-324 (1992). .
Tratschin, J.D., et al., Mol. Cell. Biol. 5:3251-3260 (1985). .
Hermanat, P.L., et al., Proc. Natl. Acad. Sci. USA 81:6466-6470
(1984). .
Lebkowski, J.S., et al., Mol. Cell. Biol. 8:3988-3996 (1988). .
Samulski, R.J., et al., EMBO J. 10:3941-3950 (1991). .
Kotkin, R.M., et al., Proc. Natl. Acad. Sci. USA 87:2211-2215
(1990). .
Pear, W.S., et al., Proc. Natl. Acad. Sci. USA 90:8392-8396, 1993).
.
Shpall, E.J., et al., Journal of Hematotherapy 3:145-147 (1994).
.
Miller, A.D., et al., Human Gene Therapy 1:5-14, (1990). .
Bahnson, A.B., et al., Gene Therapy 1(3):176-184, (1994). .
Armentano et al., J. Virol. 61:1647-1650, (1987). .
Hock et al., Blood 74:876-881, (1989). .
Berenson, R. J., et al., Blood 77(8):1717-1722, (1991). .
Bregni, M., et al., Blood 80:1418-1422, (1992). .
Nimgaonkar, M.T., et al., Gene Therapy 1:201-207, (1994). .
Barranger, J.A. et al., Intl. Pediatrics 10:5-9, (1995). .
Sanger, F., et al., Proc. Natl. Acad. Sci. USA 74:5463-5467 (1977).
.
Chen, C., et al., Mol. Cell Biol. 7:2745-2752 (1987). .
Ohashi, T., et al., J. Biol Chem. 266:3661-3667 (1991). .
Barnevald, R.A., et al., Eur. J. Biochem. 134:585-589 (1983). .
Eppig, J.J., et al., Nature 269:517-518 (1977). .
Barranger, J.A., N. Eng. J. Med. 311:1629-1631 (1984). .
Gregory, S.H., et al., J. Leukocyte Biol. 43:67-79 (1988). .
Stanley, E.R., et al., J. Biol. Chem. 252:4305-4312 (1977). .
Langley, K.E., et al., Arch. Biochem. Biophys. 295(1):21-28, 1992).
.
Danos, O., et al., Proc. Natl. Acad. Sci. USA 85:6460-6464,
1988)..
|
Primary Examiner: Guzo; David
Attorney, Agent or Firm: Baker & Botts, LLP
Parent Case Text
SPECIFICATION
This application is a continuation-in-part of our application Ser.
No. 07/904,809, filed Jun. 26, 1992, now abandoned. This invention
was made with funding from the U.S. Government, which has certain
rights therein.
Claims
We claim:
1. A human hematopoietic progenitor cell transduced with a
retroviral vector, said vector comprising and expressing a DNA
molecule that codes for glucocerebrosidase, wherein the transduced
cell provides individual with Gaucher disease biologically active
glucocerebrosidase.
2. The human hematopoietic cell of claim 1, wherein said cell is a
hematopoietic stem cell.
3. The cells of claim 1, wherein the retroviral vector is R-GC.
4. The cells of claim 1, wherein the retroviral vector is
MFG-GC.
5. The vector of claim 1, wherein the vector is R-GC.
6. The vector of claim 1, wherein the vector is MFG-GC, as
deposited with the American Type Culture Collection and assigned
accession number 75,733.
7. The cells of claim 2, wherein the cells are human CD34+
cells.
8. The cells of claim 2, wherein the retroviral vector is R-GC.
9. The cells of claim 2, wherein the retroviral vector is
MFG-GC.
10. The cells of claim 7, wherein the retroviral vector is
R-GC.
11. The cells of claim 7, wherein the retroviral vector is
MFG-GC.
12. A retroviral vector selected from the group consisting of R-GC
and MFG-GC.
13. A method for providing biologically active glucocerebrosidase
to the cell of an individual with Gaucher disease, comprising:
a) isolating autologous bone marrow from the individual with
Gaucher disease;
b) enriching the autologous bone marrow for hemapoietic progenitor
cells to obtain an enriched hematopoietic progenitor cell
population;
c) transducing the enriched progenitor cell population with a
retroviral vector that contains and expresses the
glucocerebrosidase gene; and
d) transplanting the transduced autologous progenitor cell
population into the individual with Gaucher disease so as to
provide to the individual biologically active
glucocerebrosidase.
14. The method of claim 13, in which the hematopoietic progenitor
cells are human CD34+ cells.
15. The method of claim 13, in which the retroviral vector is
R-GC.
16. The method of claim 13, in which the retroviral vector is
MFG-GC.
17. The method of claim 14, in which the retroviral vector is
R-GC.
18. The method of claim 14, in which the retroviral vector is
MFG-GC.
19. A method for providing biologically active glucocerebrosidase
to the cells of an individual with Gaucher disease, comprising:
introducing an enriched bone marrow hematopoietic progenitor cell
population into a Gaucher individual, said progenitor cell
population having been treated in vitro to insert therein a DNA
molecule encoding human glucocerebrosidase protein, said
hematopoietic progenitor cell population expressing in said Gaucher
individual biologically active glucocerebrosidase protein.
20. The method of claim 19, in which the enriched bone marrow
hemapoietic progenitor cell population comprises human CD34+
cells.
21. The method of claim 13, wherein the transducing step is
performed by centrifuging the hematopoietic progenitor cells with a
retroviral containing supernatant.
22. The method of claim 19, wherein the DNA molecule encoding human
glucocerebrosidase is inserted into the hematopoietic progenitor
cell population by centrifuging the hematopoietic progenitor cells
with a retroviral containing supernatant so as to effect
transduction of the cell population.
Description
BACKGROUND OF THE INVENTION
Gaucher disease is the name given to a group of lysosomal storage
disorders caused by mutations in the gene that codes for an enzyme
called glucocerebrosidase ("GC"). Gaucher disease is caused by
deficiency of GC as reported by Patrick, A. D., Biochem. J. 97:17C
(1965) and Brady, R. O., et al., Biochem. Biophys. Res. Commun.
18:221 (1965). All of the mutations in the gene alter the structure
and function of the enzyme which lead to an accumulation of the
undegraded glycolipid substrate glucosylceramide, also called
glucocerebroside, in cells of the reticuloendothelial system. Each
particular mutation of the human GC gene leads to a clinical
disease collectively known as Gaucher disease. These disorders are
usually classified into three types; type 1 (non-neuronopathic),
type 2 (acute neuronopathic) and type 3 (subacute neuronopathic),
the type depending on the presence and severity of neurologic
involvement. Gaucher disease is the most prevalent Jewish genetic
disease and the most common lysosomal storage disease.
GC is a monomeric, membrane-associated, hydrophobic glycoprotein
with a molecular weight of 65,000 daltons. Human GC contains 497
amino acids and is translated as a precursor protein with a 19
amino acid hydrophobic signal peptide which directs its
co-translational insertion into the lumen of the endoplasmic
reticulum-golgi-lysosome complex as reported by Erickson, A. H., et
al., J. Biol. Chem. 260: 14319 (1985). GC acts at the acidic pH of
the lysosome to hydrolyze beta-glucosidic linkages in complex
lipids ubiquitously found in all membranes to form the byproducts
of glucose and ceramide. The catalytic activity of GC is increased
in vitro by detergents, lipids, and in vivo by a naturally
occurring activator known as sphingolipid activator protein-2
(SAP-2 or saposin C). See, Ho, M. W., et al., Proc. Natl. Acad.
Sci. USA 68:2810 (1971); and O'Brian, J. S., et al., Science
241:1098 (1988).
Human GC cDNA was first cloned as described by Ginns, E. I., et
al., Biochem. Biophys. Res. Commun. 123:574 (1984). Subsequent
characterizations of other GC cDNA clones by, for example, Sorge,
J., et al., Proc. Nat. Acad. Sci. USA 82:7289 (1985) and Tsuji, S.,
et al., J. Biol. Chem. 261:50 (1986), have led to the elucidation
of the complete nucleotide sequence of human GC. As reported by
Ginns, E. I., et al., Proc. Nat. Acad. Sci. USA 82:7101 (1985), the
GC gene was localized to human chromosome lq21 by in situ
hybridization. Tsuji, S., et al., New Enql. J. Med. 316:570 (1987),
have shown that the GC gene comprises 11 exons and 10 introns
spanning approximately 7 Kb.
While more than twenty mutations in the human GC gene are known,
only two are common. See, Tsuji, S., et al., Proc. Natl. Acad. Sci.
USA 85:2349 (1988). The two common mutations account for
approximately 70% of the mutant alleles, as reported by Firon, N.,
et al., Am. J. Hum. Genet. 46:527 (1990). Mutant GC genes code for
aberrant proteins that are either catalytically altered or unstable
and rapidly disappear from the cell.
Although GC is deficient in all of a subject's cells, for unknown
reasons, the accumulation of the substrate glucosylceramide occurs
virtually only in macrophages. Gaucher disease is unique among
lysosomal storage disorders for this reason, that is, causing
storage within only one cell type. This characteristic of the
pathobiology of the disease has led to the development of two
successful treatment strategies based on correcting the enzyme
deficiency in macrophages.
The first of the two Gaucher disease treatments based on this
strategy is allogeneic bone marrow transplantation, which results
in the repopulation of affected tissues with enzyme-competent
macrophages. See, Rappeport, J. M., et al., Birth Defects: Original
Article Series 22,1:101 (1986). The second approach to treatment
which has resulted in clinical improvement in Gaucher disease
patients is macrophage-targeted enzyme replacement. This treatment
takes advantage of naturally occurring mannose receptors on
macrophages and the exposition of accessible mannose receptors in
the oligosaccharides of glucocerebrosidase to efficiently deliver
the enzyme to macrophages. See, Barranger, J. A., et al., Japanese
J. of Inher. Met. Disease 51:45 (1989); Takasaki, S., et al., J.
Biol. Chem. 259:10112 (1984); and Furbish, F. S., et al., Biochem.
Biophys. Acta. 673:425 (1981). While both of these approaches to
treating Gaucher disease are important because they provide some
means of therapy where none previously existed, both approaches
have significant limitations. Allogeneic bone marrow
transplantation has associated with it morbidity and mortality
risks that are unacceptable for many patients. Further, HLA matched
bone marrow donors do not exist for the majority of patients. As
for macrophage-targeted enzyme replacement, it is currently an
expensive and life-long therapy; thus, it should be reserved for
only the most severely ill patients.
Despite the limitations of these two therapies, their successes
have demonstrated that enzymatic correction of only one cell type,
the macrophage, results in effective therapy for Gaucher disease.
From the point of view of developing somatic cell gene therapy for
Gaucher disease, the fact that marrow transplantation is effective
demonstrates that bone marrow stem cells are an appropriate target
cell to which to transfer a "therapeutic" gene. Further, because of
the pivotal role of macrophages in Gaucher disease, alternative
target cells to be considered for gene transfer are the committed
macrophage precursors, peripheral blood monocytes, or cultures of
bone marrow which are capable of producing macrophage
precursors.
To be an effective permanent treatment for any disease capable of
being treated by gene therapy, the transfer and sustained
expression of genes in cells important to the pathogenesis of the
particular disease is required. Sufficient expression of a
transduced GC gene in the progeny of pluripotent bone marrow stem
cells would likely correct the deficiency of the enzyme in all cell
series including monocytes/macrophages. Experience from allogeneic
marrow transplantation and macrophage-targeted enzyme replacement
supports the idea that gene therapy would provide a cure provided
adequate expression of the GC gene were achieved in a sufficient
number of macrophages.
Much experience has been gained recently to evaluate the efficiency
of gene transfer and expression in bone marrow stem cells using
replication defective retroviral vectors. See, e.g., Miller, A. D.,
Blood 76:2 (1990) and Miller, D. G., et al., Mol. Coll. Biol.
10:4239 (1990). Most of the studies of retroviral vectors have been
conducted in the mouse model of bone marrow transplantation. Recent
data show that 10-20% of stem cells can be transduced and survive
to repopulate marrow. See, Bodine, D. M., et al., Exp. Hematol.
19:206 (1991). Many fewer studies have been conducted in larger
animals and, at the present time, experimental conditions have not
yet been fully optimized. For the mouse model, critical parameters
for efficient retroviral gene transfer and repopulation of bone
marrow include high titer virus producer cell lines (VPL),
pretreatment of mice with 5-fluorouracil (5-FU) to initiate stem
cell cycling, pre-culture of bone marrow with growth factors
including IL-3 and IL-6 and stem cell factor (SCF). Several studies
have shown that the GC gene can be transferred to murine bone
marrow stem cells and their progeny, but until recently none had
demonstrated expression of enzymatic activity in macrophages in
vivo. See, Nolta, J. A., et al., Blood 75:75 (1990) and Correll, P.
H., et al., Proc. Natl. Acad. Sci. USA 86:8912 (1989).
Retroviral vectors for use in gene therapy require dividing cells
in order to integrate and they have a small, but finite chance of
interrupting an essential gene or altering expression of a gene
proximate to the site of integration of the retroviral provirus.
However, the proviral integration may be preferentially directed to
transcriptionally active regions of the genome as described by
Scherdin, U., et al., J. Virol. 64; 2:907 (1990). These
requirements of retroviral vectors are believed to contribute to
the small number of stem cells that can be transduced, since only a
portion of the stem cell population is cycling even under optimal
experimental conditions. See, McLaughlin, S. K., et al., J. Virol.
65:1963 (1991).
For these and other reasons, the small helper-dependent human DNA
parvovirus known as adeno-associated virus ("AAV") has recently
received attention as a vector that could be useful for gene
therapy. See, Hunter, L. A., et al., J. Virol. 66:317 (1992);
Tratschin, J. D., et al., Mol. Cell. Biol. 5:3251 (1985); Hermanat,
P. L., et al., Proc. Natl. Acad. Sci. USA 81:6466 (1984); and
Lebkaski, J. S., et al., Mol. Cell. Biol. 8:3988 (1988). A lytic
growth cycle for wild AAV requires infection with a helper virus,
e.g., adenovirus type 5. In the absence of helper virus, AAV
integrates into the host genome by hybridization between the AAV
terminal repeats (TR) and host sequences in a stable manner thereby
establishing a permanent latent infection. In human cells, this
integration occurs preferentially in a single silent site in human
chromosome 19. See, Samulski, R. J., et al., EMBO J. 10:3941
(1991); Kotkin, R. M., et al., Proc. Natl. Acad. Sci. USA 87:2211
(1990).
In order to develop gene therapy that would be useful for the
treatment of Gaucher disease as well as other hematopoietic
disorders, new vectors are needed that allow efficient gene
transfer to stem cells and at the same time direct expression of
the transferred gene. Thus far there has not been a vector that can
transduce and sustain the expression of the human GC gene in bone
marrow stem cells and their progeny. Accordingly there is a need
for such a vector which could be used in gene therapy for Gaucher
disease.
SUMMARY OF THE INVENTION
Accordingly, it is a primary object of the present invention to
provide replication defective retroviral vectors that are able to
transduce and sustain the expression of the human GC gene in
mammalian bone marrow stem cells and their progeny.
Another object of the present invention is to provide replication
defective retroviral vectors capable of transducing and expressing
GC activity in committed macrophage precursors.
Still another object of the present invention is to provide
replication defective retroviral vectors capable of transducing and
expressing GC activity in macrophages.
These and other objects of the present invention are achieved by
one or more of the following embodiments.
In another aspect, the invention features the replication defective
retroviral vectors MFG-GC and R-GC.
In still another aspect the invention features mammalian bone
marrow stem cells transduced with MFG-GC or R-GC.
In yet another aspect the invention features mammalian macrophage
precursors transduced with MFG-GC or R-GC.
In preferred embodiments, the invention features a method of
treating Gaucher disease in a patient with Gaucher disease,
comprising administering to said patient a therapeutically
effective amount of mammalian bone marrow cells or mammalian
macrophages transduced with MFG-GC or R-GC, such that the
transduced cells are stably maintained in the patient and express
therapeutic levels of glucocerebrosidase.
In other preferred embodiments, the invention features a method of
treating Gaucher disease in a human patient, comprising
administering a therapeutically effective amount of human bone
marrow cells or human macrophages transduced with MFG-GC or R-GC,
such that the transduced cells are stably maintained in the patient
and express therapeutic levels of glucocerebrosidase.
In another aspect, the invention features replication defective
retroviral vectors containing the GC gene and capable of
transducing and long term expression of the GC gene in ex vivo or
in vivo gene therapy treatment.
Other features and advantages of the invention will be apparent
from the following description of the preferred embodiment, and
from the claims.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention may be better understood by reference to the
following figures.
FIG. 1 Restriction site map of the MFG vector.
FIG. 2 Shows the strategy for the assembly of the MFG-GC vector of
the present invention.
FIG. 3 Shows the structure of the MFG-GC provirus.
FIG. 4 Photograph of a Western blot of immunoreactive human GC
protein resulting from infection of NIH 3T3 murine fibroblast cells
with MFG-GC- described in Table 1 (lanes 1-5) as compared to
uninfected NIH 3T3 ells (lane 6) and human fibroblast 0637D cells
(lane 7).
FIG. 5 Photograph of a Southern blot of NA from MFG-GC infected NIH
3T3 fibroblasts and probed with human .sup.32 P-GC-cDNA. Lanes 1-3
are controls containing 1 pg, 5 pg, and 10 pg of MFG-GC,
respectively, and lanes 4-8 contain DNA from the MFG-GC infected
3T3 cells described in Table 1.
FIG. 6 Chart showing results of Southern nd Western analyses, and
enzymatic activity of cells infected by the MFG-GC and N2-SV-GC
vectors. Results of assays on 3T3 cells, spleen colonies at day 12
(CFU-S.sub.12) and long term bone marrow cultures (LTBMC) are
shown.
FIG. 7 Graph showing the results of enzymatic analyses of spleen
colonies from three mice (A1, A2, and A3) receiving marrow infected
by the MFG-GC vector as compared to a colony in the spleen of a
mouse transplanted with normal bone marrow.
FIG. 8 Photograph of normal mouse leukocytes immunocytochemically
stained for human GC protein with monoclonal antibody 8E4.
FIG. 9 Photograph of mouse leukocytes immunocytochemically stained
for human GC. The cytospin preparation was made from the buffy coat
of whole blood drawn from a mouse reconstituted with bone marrow
infected with the MFG-GC vector 5 months previously.
FIG. 10 Photograph of a higher power view of FIG. 9.
FIG. 11 Table comparing MFG-GC and N2-SV-GC retroviral vectors by
Southern and Western blots, enzymatic activity in 3T3 cells, spleen
colonies, LTBMC, and tissues of mice surviving longer than four
months after reconstitution with bone marrow infected by
co-cultivation with each of the two viral producer lines.
FIG. 12 Graph showing GC activities in the tissues (liver, lung,
spleen, lymph node (LN), and bone marrow (BM)) of four mice
surviving longer than four months after reconstitution with bone
marrow infected by co-cultivation with MFG-GC, with the control
being the activity present in the respective tissues of normal
mice.
FIG. 13 Photograph of latex bead ingestion by macrophages grown
from the bone marrow of long term reconstituted mice that had been
rescued with marrow infected with MFG-GC.
FIG. 14 Photograph of macrophages grown from long term
reconstituted mice rescued with marrow infected with MFG-GC five
months previously that were immunocytochemically stained for human
GC.
FIG. 15 Photograph of macrophages cultured from normal murine bone
marrow that were immunocyto-chemically stained for human GC.
FIG. 16 Photograph of macrophages grown from long term
reconstituted mice (>5 mos.) rescued with marrow infected with
N2-SV-GC that were immunocytochemically stained for human GC.
FIG. 17A Graph showing results of enzymatic analyses of macrophages
cultured from explanted marrow of eight long term reconstituted
mice rescued with marrow infected with MFG-GC (lanes 1-8), and
macrophages cultured from a normal control mouse (lane 9).
FIG. 17B Photograph of a Western blot of immunoreactive human GC
protein expressed in extracts from macrophages cultured from
explanted marrow of a long term reconstituted mouse rescued with
marrow infected with MFG-GC (lane 1) and in extracts from a control
mouse macrophage culture (lane 2).
FIG. 17C Photograph of a Southern blot of DNA from macrophages
cultured from explanted marrow of a long term reconstituted mouse
rescued with marrow infected with MFG-GC (lane 1). Lane 2 is a
normal mouse control, and lanes 3-5 contain standards for
estimating copy number representing 0.2, 0.5, and 1.0 copies/cell,
respectively. The DNAs were probed with human 32P-GC-cDNA.
FIG. 18A Photograph of Southern blots of DNA from 27 individual
secondary CFU-S.sub.12 spleen colonies from three mice that
received bone marrow transplants from primary long term
reconstituted mice and probed with full length human GC cDNA. Lanes
1-10 contain DNA from spleen colonies isolated from mouse #26;
lanes 11-19 contain DNA from spleen colonies isolated from mouse
#27; lanes 20-27 contain DNA from spleen colonies isolated from
mouse #28; lane 28 contains genomic DNA isolated form an
age-related control; and lanes 29-31 contain standards for
estimating copy number (0.2, 0.5, and 1.0 copies/cell,
respectively).
FIG. 18B Graph showing results of enzymatic analyses of 27
individual secondary CFU-S.sub.12 spleen colonies from three mice
that received bone marrow transplants from primary long term
reconstituted mice. Lane 1 represents GC activity of an age-related
control; and lanes 2-10, 11-19, and 20-27 represent, respectively,
spleen colonies from mouse #26, #27, and #28.
FIG. 19 (Parts A-B) Graph of the GC activity of normal human and
Gaucher patient macrophages in culture before (A) and after (B)
infection with the MFG-GC vector expressed as nanomoles of 4MU
released per hour per mg of cell protein.
FIG. 20 (Parts A-B) Results of flow cytometry of hematopoeitic
cells using anti-CD34+ monoclonal antibody to enrich for CD34+
cells (A). Post-enrichment analysis shows 87.6% CD34+ cells
(B).
FIG. 21 Clonogenic assays of pre- and post-enriched samples to
determine numbers of colony forming units (CFU) demonstrated
increased numbers of CFU with CD34+ cells compared to the
pre-enriched samples.
FIG. 22 CD34+ PB cells were exposed to viral supernatants 6 times
over a period of 5 days beginning three days after incubation in
mixtures of cytokines as indicated. Most of the cells were
harvested one day after the final infection and were analyzed for
enzyme activity. Values are the means for duplicate pellets.
FIG. 23 Samples of infected and control CD34+ cells were harvested
and analyzed for GC enzyme activity at the times indicated, while
the remaining cells were further expanded in culture over a period
of three weeks following infection.
FIG. 24 Following infection and expansion into medium containing
rrSCF, cell growth was observed to diminish. Aliquots of two
samples were transferred to two dishes each, and to one dish from
each sample was added hrSCF. These samples were incubated for an
additional 6 days prior to counting cell numbers in a
hemacytometer.
FIG. 25 CD34+ PB cells were infected and expanded under different
conditions as described in the test. One day after the final
infection, cells were harvested and were analyzed for GC specific
activity. Error bars indicate mean +/-se (n=3).
FIG. 26 Summary of experiments 1-5, showing enzyme activity in
CD34+ PB cells following infection with MFG-GC containing
supernatants. See text for details. Error bars indicate mean +/-se
(n=4 or more cell pellets analyzed, except Experiment 5; control
n=2).
FIG. 27 Southern blot hybridization of Sst 1 digests of DNA from
infected and noninfected CD34+ enriched PB cells (Experiment 5).
The 4.3 kb fragment (apparent mol. wt.) in the vector, to which the
1.8 kb GC cDNA probe hybridizes, as well as the 6.4 kb Sst 1
fragment of the normal GC gene and the pseudogene are shown. Copy
number controls are shown, prepared from known amounts of DNA from
a murine cell line containing 5 copies per cell.
FIG. 28 GC activity from CD34+ cells isolated from normal bone
marrow and cord blood transduced with MFG-GC relative to control
(noninfected) cells.
FIG. 29 Human bone marrow CD34+ cells were infected by daily
exposure to virus-containing supernatant for 4 days. Prior to the
second and each subsequent infection, an aliquot of cells was
transferred to a separate dish without further exposure to virus.
After the infection period, cells were expanded and harvested for
analysis of enzyme expression.
FIG. 30 GC activity of MFG-GC transduced CD34+ cells from Gaucher
bone marrow relative to noninfected control cells.
FIG. 31 (Parts A-B) Immunocytochemical staining for human
glucocerebrosidase (GC). Cytospin preparations were stained using
the monoclonal antibody 8E4. Normal human GC stains red in Gaucher
marrow transduced with the MFG-GC vector (A), while no staining
occurred in nontransduced Gaucher marrow cells (B).
FIG. 32 Structure of R-GC retroviral vector.
FIG. 33 Diagrammatic representation of cross-infection strategy for
deriving R-GC producer cell clones.
FIG. 34 R-GC supernatants assayed on 3T3 target cells for GC
activity to select the highest titer amphotropic producer
(LM30).
FIG. 35 The GC activity conferred on target cells infected by LM30
supernatants collected periodically over a continuous period. The
relative titer of the LM30 producer was maintained over time for as
long as 2 years.
FIG. 36 R-GC and MFG-GC were obtained from the YCRIP producer cell
lines, LM30 and cc-2, respectively. The cc-2 supernatant was from
the highest-titer pooled batch, P3.times.8--Aug. 24, 1993. The LM30
supes were from the highest-titer individual supernatants available
at the time: PB9=LM30 3.times.6--Dec. 20, 1993, and PB10=LM30b
3.times.9--Feb. 7, 1994. The cells from PB9 were 5% CD34+ by FACS
analysis, and were infected with only one addition of supernatant.
The cells from PB10 were more highly enriched for CD34+
(.about.40%), and were infected with three exposures to fresh
supernatant over 1.5 days. Both protocols included a 24 hour
prestimulation in medium containing IL-3, IL-6, and SCF. Results
for enzyme activity are expressed relative to mock-infected control
cells exposed to 10%CS/DMEM and cultured in an identical manner to
the infected cells in each experiment. Noninfected control
activities were 165 and 44 U/mg for PB9 and PB10, respectively.
FIG. 37 The GC activity in 8 separate samples of R-GC transduced
CD34+ cells from 2 Gaucher patients, in which all evidenced the
restoration of GC activity to at least that of normal cells.
NI=noninfected. GD=Gaucher disease.
FIG. 38 Correction of GC deficiency in mobilized cells obtained
from a single patient with Gaucher disease. GC activity is shown as
a function of days post-transduction. NI=noninfected. GD=Gaucher
disease.
FIG. 39 Titers of cryogenically preserved vials of the
YCRIP--MFG-GC producer (cc2) were determined by infection of 3T3
targets and conversion of enzyme activity to copies/ml through
correlation of Southern blot hybridization intensity as described
elsewhere (Bahnson, A. B., et al., Gene Therapy 1(3):176-184,
1994). Bars indicate the mean and standard error for n=2 to 4 supes
assayed.
FIG. 40 Flow sheet for clinical trial of gene therapy for Gaucher
disease using the retroviral vector R-GC.
DETAILED DESCRIPTION OF THE INVENTION
Gene therapy for Gaucher disease may be accomplished using a viral
vector to introduce the GC gene into hematopoietic cells to
genetically correct the inability to synthesize functional GC
enzymatic activity. In a particular embodiment, a retroviral vector
may be used. Vectors which may be used include, but are not
limited, to N2 (Armentano et al., J. Virol. 61:1647-1650), 1987),
LNL6 (Hock et al., Blood 74:876-881, 1989) and MFG (published PCT
application WO92/07943, published May 14, 1992).
The retroviral vector is engineered to contain the human GC gene,
using standard recombinant DNA techniques known to those skilled in
the art (Current Protocols in Molecular Biology. Ausubel, F. et
al., eds., Wiley and Sons, New York 1995). The human GC gene can be
cloned into an MFG vector to create MFG-GC, in one embodiment, or
to create the R-GC vector, in another embodiment. The embodiment,
R-GC, is a vector that is derived from MFG-GC, but which has an
additional safety feature--an insertion of a SacII linker into the
gag sequence to cause a frameshift that prevents the synthesis of
gag-related peptides. In other embodiments, safer retroviral
vectors may be produced by altering the gag coding sequence with
the insertion of stop codons or by mutation of the start codon in
the sequence, so that gag-related peptides are not synthesized.
The vectors of the invention may be used in ex vivo or in vivo gene
therapy.
Stocks of the retroviral vector may be produced from packaging cell
lines that provide the necessary structural proteins (Miller, A.
D., Human Gene Therapy 1:5-14, 1990) and enable a
replication--defective vector stock to be produced.
Quantitation of the vector stock is performed by using standard
assays, including plaque formation on suitable target cells, virus
protein assays (e.g., p.24), or measurement of the activity of a
transduced gene (i.e., GC activity). High titer producer clones may
be chosen for use as a source material.
Gene Transfer and Expression in CD34+ Cells From Blood
CD34+ cells contain a percentage of pluripotent stem cells capable
of reconstituting the bone marrow in primates and man (Berenson, R.
J., et al., Blood 77(8):1717-1722, 1991; Shpall, E., et al., Blood
82:321, 1993). Studies have shown that the human peripheral blood
CD34+ cell (PBSC) can be transduced by retroviral vectors (Bregni,
M., et al., Blood 80:1418-1422, 1992). Because of the ability to
concentrate CD34+ cells in a small volume, the logistics of ex vivo
transduction become greatly simplified. Furthermore, the ratio of
virus to stem cell in infection protocols can be enhanced, leading
to more efficient transduction. Because of this economy and because
the bone marrow of patients with Gaucher disease are frequently
"packed" (full of cellular material and scar; they frequently can
not be aspirated), the PBSC are particularly ideal cells for gene
transfer/therapy studies.
For gene therapy treatment, granulocyte colony stimulating factor
(G-CSF) may be administered to an individual in order to increase
the number of CD34+ cells circulating in the blood. In an ex vivo
protocol, leukopheresis may be used to collect the white blood
cells. Density gradient separation of peripheral blood mononuclear
cells (MNC) may be performed. CD34+ cells may be enriched from the
MNC by immunoaffinity purification. This enrichment generates a
stem cell population that optimizes the permanence of gene transfer
due to its pluripotent capability.
The enriched CD34+ cell population may be transduced with the
supernatant from an amphotropic producer cell line that generates a
high titer retroviral vector stock. In a preferred aspect of the
invention, R-GC is the retroviral vector that is used to transduce
the CD34+ cells, for gene therapy of Gaucher patients.
The transduced CD34+ cells may be assayed for transduction
efficiency by techniques known to those skilled in the art,
including PCR identification of the GC gene, Southern blot
determination of gene copy number, GC enzymatic assay, and
immunocytochemistry to detect the GC protein (Bahnson, A. B., et
al., Gene Therapy 1:176-184, 1994; Nimgaonkar, M. T., et al., Gene
Therapy 1:1-7, 1994; Barranger, J. A. et al., Intl. Pediatrics
10:5-9, 1995).
Upon the establishment of an autologous R-GC-transduced CD34+ cell
population, the cells are transfused back into a Gaucher patient.
Clinical progress is monitored by ongoing analysis of peripheral
blood leukocytes (PBL) to determine the presence of the GC gene
(PCR, Southern blotting) and the activity of the GC enzyme. Other
clinical parameters that may be used to assess patient status
include routine blood profiling including hemoglobin level and
platelet count, size of liver and spleen, biopsy of bone marrow and
x-rays of long bones and the pelvis (see FIG. 40).
EXAMPLE 1
Construction of MFG-GC Vector
The starting vector (MFG) from which a replication defective
retroviral vector containing the human GC gene (MFG-GC) was
constructed as described in published PCT application WO92/07943,
published May 14, 1992. (ATCC accession number ATCC 68,754) The
physical structure of the MFG vector is shown in FIG. 1. The
features of the present MFG-GC vector deposited at the American
Type Culture Collection, 12301 Parklawn Drive, Rockville, Md. 20852
on Apr. 8, 1994 and given the (ATCC accession number 75733) are
that the GC cDNA is transcribed by the retrovirus long terminal
repeats ("LTR") and the position of the start codon of the GC cDNA
is placed exactly at the start codon of the deleted envelope
protein gene. No internal promoter or dominant selectable marker is
included in the construct. Additionally, there is a Sac II
polylinker at the ATG of gag which may further reduce the
possibility of the replication defective vector recombining to
produce replication competent virus. GC cDNA is inserted into an
Nco I/Bam HI site.
The strategy for the construction of the MFG-GC vector is shown in
FIG. 2. In order to create an Nco I site in the position of the
start codon (ATG) of GC cDNA, polymerase chain reaction (PCR) was
carried out with the primers shown in FIG. 2. The sense primer was
made with a one base pair mismatch to create an Nco I site. The
antisense primer was located downstream of a Hind III site. PCR was
carried out using a thermal cycler (Perkin Elmer, Norwalk, Conn.)
according to the manufacturer's protocols. The resulting PCR
product was then cut with Nco I and Hind III and a 60 bp fragment
was isolated using 4% NuSieve.TM. agarose gel (FMC Bio Products,
Rockland, Me.). On the 3' side of the GC cDNA, an Eco RI fragment
of GC cDNA was isolated and the terminus was filled in with dNTP's
by the Klenow fragment (Boehringer Mannheim, Gmblt, Germany) to
create a blunt end. A Bcl I linker was ligated to this fragment in
the usual manner and was digested with Hind III and Bcl I. The
resulting 1.7 Kb Hind III/Bcl I fragment was isolated from 1% low
melting point agarose gel. The 60 bp Nco I/Hind III fragment and
1.7 Kb Hind III/Bcl I fragment were then ligated to the Nco I/Bam
HI site in the MFG vector. The construction was confirmed by DNA
sequencing according to the method of Sanger, F., et al. Proc.
Natl. Acad. Sci. USA 74:5463 (1977), which revealed no PCR errors.
The structure of the MFG-GC provirus is shown in FIG. 3.
Alternative constructions of replication defective retroviral
vectors containing the GC gene and which are capable of transducing
and long term expression of the GC gene are also within the scope
of the present invention. One example incorporates a Neo.RTM. gene
into the MFG-GC vector for the purposes of selection of transduced
cells and may be used to select an enriched population of human
cells carrying the therapeutic GC gene.
Isolation of Virus Producing Cells and Titering
The psi-cre producer line, as described by Danos, O., et al., Proc.
Natl. Acad. Sci. USA 85 6460 (1988), and NIH 3T3 murine fibroblast
cells (ATCC, Rockville, Md.) were cultivated in Dulbecco's modified
Eagle's medium (DMEM) (Gibco, Grand Island, N.Y.), supplemented
with 4.5 g/l glucose, 3.7 g/l NaHCO.sub.3, 10% heat inactivated
calf serum, 100 U/ml penicillin (Gibco), and 200 .mu.g/ml
L-glutamine (Gibco). MFG-GC was co-transfected with pSV2Neo to
Psi-cre, because no dominant selectable marker is present in this
construct. After selection by the neomycin analog Geneticin G418
(Gibco) (400 .mu.g/ml), 30 clones were isolated and grown to near
confluence in 100 mm dishes. The culture medium of these clones was
used as the virus source. As a target cell for titering, NIH 3T3
murine fibroblast cells were used. The virus containing medium was
added to 3T3 cells and grown to a density of 4.times.10.sup.4 in 6
well dishes. Polybrene (Sigma, St. Louis, Mo.) was also added to
the medium at a concentration of 8 gg/ml to facilitate entry of the
virus to the cells. After a 2 hour incubation at 37.degree. C., the
virus containing medium was removed and fresh medium without virus
was added. Following a 48 hour incubation at 37.degree. C., the
cells were harvested and analyzed by measurement of GC activity,
Western blots, and Southern blots.
To provide a comparison with the MGF-GC vector of the present
invention, another retroviral vector, N2-SV-GC was constructed and
used to infect 3T3 cells and bone marrow. The N2-SV-GC vector was
constructed by inserting a normal GC cDNA under control of the SV40
early region promoter into the Moloney murine leukemia
virus-derived N2 vector, as described by Nolta, J. A., et al.,
Blood 75, 3:787 (1990). In order to isolate virus producing cells
and titer the virus, N2-SV-GC constructs were transfected to the
ecotropic packaging line, psi-cre, using the calcium phosphate
co-precipitation method of Chen, C., et al., Mol. Cell Biol. 7:2745
(1987). NIH 3T3 target cells infected by N2-SV-GC were selected by
G418 (400 mg/ml). After 2 weeks, the G418 resistant colonies were
counted to determine the virus titer. Southern blots of
non-selected cells were performed for comparisons of titers with
the MFG-GC vector.
EXAMPLE 2
TRANSFER OF HUMAN GC GENE TO 3T3 CELLS BY MFG-GC VECTOR
GC Enzymatic Assay
NIH 3T3 murine fibroblasts were infected with either MFG-GC or
N2-SV-GC virus-containing supernatants as described above and
selected in 0.5 mg/ml G418 for 1 week. Fibroblasts for GC enzymatic
assay were trypsinized, washed twice with Hank's Buffered saline
solution (HBSS), and lysed in "sonication buffer" (50 mmol/l
potassium phosphate buffer, pH 6.5 with 0.25% Triton X-100).
Samples were sonicated for two 30-second pulses at 40 watt-seconds
and centrifuged at 12,000 RPM for 2 minutes at 4.degree. C. The
protein concentrations of the cleared sonicates were determined
with the Pierce BSA kit (Pierce, Rockford, Ill.) using bovine serum
albumin (BSA) for construction of a standard curve.
GC enzymatic activity was measured by the method of Ohashi, T., et
al., J. Biol Chem. 266:3661 (1991), the disclosure of which is
incorporated herein by reference, with the synthetic fluorogenic
substrate 4-methylumbelliferyl-.beta.-D-glucopyranoside (4MU-glc)
(Sigma, St. Louis, Mo.). The reaction mixture (200 .mu.l) contained
5 mM 4MU-glc, 0.1M citrate phosphate buffer (pH 5.4), 2.1 mM Triton
X-100, 0.1% bovine serum albumin, and 3.5 mM sodium taurocholate.
After incubation of each cell lysate with the substrate for 30
minutes at 37.degree. C., reactions were terminated by the addition
of 3.8 ml of 0.17M glycine carbonate buffer (pH 10.4). The
4-methylumbelliferone (4MU) that had formed was measured
fluorometrically. Units of GC activity were expressed as nanomoles
of 4MU substrate produced/hour/mg protein. All values represent the
average of duplicate measurements.
The enzymatic activity of MFG-GC infected NIH 3T3 cells with the
highest activity (best 5 out of 20) are shown in Table 1 below. All
of these cells exhibit approximately 5 to 10 times higher activity
than non-infected NIH 3T3 cells.
TABLE 1 ______________________________________ Cells GC Activity
______________________________________ GC# 1 1392 GC# 4 2356 GC# 5
1364 GC# 25 1261 GC# 31 1841 NIH 3T3 250 .+-. 9*
______________________________________ Activity was expressed as
nmol/h/mg protein *: Mean .+-. SD of three separate
determinations
Western Blot Analysis
Cells and tissues of interest were lysed in 50 mM potassium
phosphate buffer (pH 6.5) containing 0.25% Triton X-100 either by
sonication or homogenization using 5 a Dounce type homogenizer. 100
.mu.g of protein was applied to a 7.5% SDS-polyacrylamide gel and
electroblotted onto a nitrocellulose membrane. The human GC protein
was detected by immunostaining using monoclonal antibody 8E4 and
alkaline phosphatase conjugated anti-mouse Ig-G goat antibody
(BioRad, Richmond, Calif.). The 8E4 monoclonal antibody is specific
for human GC and does not cross react with endogenous mouse GC.
See, Barnevald, R. A., et al., Eur. J. Biochem. 134:585 (1983), and
Ohashi, T., et al., J. Biol. Chem. 266:3661 (1991).
The results of the Western blot analyses are shown in FIG. 4. Lanes
1-5 contained immunoreactive expression product from the five
samples of MFG-GC infected NIH 3T3 cells shown in Table 1. Lane 6
contained non-infected NIH 3T3 cells and lane 7 contained a human
fibroblast line, 0637D (ATCC accession number as a control which is
an SV40 Ori transformed normal human fibroblast that carries one
copy of the GC gene pet cell. Because of the very strong signal for
GC in the 3T3 target cells, the Western analysis was adjusted so
that the 0637D signal was just at the limit of detection in order
to estimate expression in the target cells. The expression in the
targets was estimated to be about 5-10 times that observed in the
control. The blots showed the expressed protein was of the expected
size range (59-66 Kd).
Southern Blot Analysis
Southern blot analyses of DNA from MFG-GC infected cells were
carried out as follows. Genomic DNA was extracted by proteinase K
treatment, phenol extraction and ethanol precipitation, according
to methods known in the art. The DNA was digested with Sst I which
cut in both LTR regions in the constructs and produced a 4.3 Kb and
4.9 Kb band from the MFG-GC and N2-SV-GC vectors, respectively. The
digested DNA was separated in 1% agarose and blotted onto a
nitrocellulose membrane. Known amounts of plasmid DNA from MFG-GC
and N2-SV-GC were run on the same gel and the vector copy number
per genome was estimated by comparison of band intensities. Human
GC cDNA was used as the probe. Hybridization and washing were
performed using standard techniques. See, Ausubel, F. M., et al.,
Eds. Current Protocols in Molecular Biology Wiley-Interscience, New
York (1995).
FIG. 5 shows the results of Southern blot analyses performed as
described above with the same five MFG-GC infected samples used in
the Western blots (lane 4-GC #31; lane 5-GC #25; lane 6-GC #5; lane
7-GC #4; lane 8-GC #1), and MFG-GC standards in varying amounts
(lane 1=2 pg; lane 2=5 pg; and lane 3=10 pg). The results reveal
approximately 1-2 copies of GC per cell. This expression correlates
well with the results of the Western blots described above.
FIG. 6 shows the results of Southern and Western blot analyses and
an enzymatic assay for 3T3 cells infected by MFG-GC and N2-SV-GC
showing the enhanced expression of GC for cells infected by MFG-GC
as compared to N2-SV-GC.
In the following example, long term bone marrow cultures (LTBMC)
were established in order to study gene transfer by MFG-GC and
expression of the GC gene.
EXAMPLE 3
TRANSDUCTION AND EXPRESSION OF THE HUMAN GC GENE IN LTBMC
Establishment and Growth of Murine LTBMC
Bone marrow was harvested from 6- to 8-week-old C57BL/6J (B6) mice
(Jackson Laboratories, Bar Harbor, Me.) by flushing the marrow
cavities of the femurs and tibiae from the donors with cold RPMI
medium containing penicillin (100 U/ml) and streptomycin (100
.mu.glml) (Gibco). The cells were washed once with RPMI-10% FCS,
resuspended in Iscove's modified Dulbecco's medium (IMDM) (Gibco),
and viable cell numbers were determined with trypan blue. Prior to
placing the marrow into culture, it was infected by either the
MFG-GC or N2-SV-GC vector as described below in Example 4 in the
protocol for infecting marrow to be used for bone marrow
transplantation. To establish LTBMC, 1.times.10.sup.7 nucleated
cells were placed into T25 vented filter-top flasks (Costar,
Cambridge, Mass.) in complete bone marrow medium (CBMM). CBMM is
IMDM, 30% HI-FCS (Gibco); 1% deionized bovine serum albumin (BSA)
(Sigma); 106 mol/l hydrocortisone (Abbott Labs, N. Chicago, Ill.);
1.times.10.sup.4 mol/l 2-mercaptoethanol (Sigma); 2 mmol/l
glutamine (Gibco); 100 U/ml penicillin (Gibco), 100 .mu.g1m1
streptomycin (Gibco); and hematopoietic growth factors provided as
5% (vol/vol) of IMDM-20 conditioned by WEHI-3B (WEHI-CM) (gift of
Dr. Sallie Boggs, University of Pittsburgh, Pittsburgh, Pa.) and 5%
(vol/vol) of IMDM-20 conditioned for 7 days by murine spleen cells
stimulated with 2.5 .mu.g/ml poke weed mitogen (PWM-CM) (Gift of
Dr. Sallie Boggs). To maintain the LTBMC, half of the culture
medium and nonadherent cells were replaced with fresh CBMM every 5
to 7 days. These LTBMC were maintained for over 10 months with
sustained production of hematopoietic progenitors (assayed in a
CFU-GEMM colony assay) and mature blood cells (determined by
Wright-Geimsa staining (Fisher, Pittsburgh, Pa.) of cytocentrifuge
preparations).
Analysis of LTBMC Infected by MFG-GC
Southern blot, Western blot, and enzymatic analyses of murine LTBMC
were performed and the results are summarized in FIG. 6. It is
shown that the MFG-GC vector was capable of infecting LTBMC very
efficiently and allowed efficient transcription and translation of
the GC MRNA. The Southern analyses show that the infection of these
various cells approached 100%. The activity of the enzyme in
transduced cells was approximately five to thirty (5-30) fold above
the background and continued at that level for more than 5 months
in culture.
In the following example the ability of MFG-GC and N2-SV-GC to
infect and be expressed in CFU-S was compared.
EXAMPLE 4
Analysis of CFU-S for Transduction Efficiency and Expression of the
Human GC Gene
Lethally irradiated mice that received 1.times.10.sup.5 donor bone
marrow cells infected with MFG-GC or N2-SV-GC were studied 12 days
after bone marrow transplantation for spleen colonies as described
below in Example 5. All animals survived the transplantation. Each
mouse had a spleen that was 8-10 times the weight of a normal
spleen and was essentially replaced by confluent colonies of bone
marrow origin.
A piece of spleen containing at least one colony was prepared from
each mouse. These spleen fragments were analyzed by Southern
blotting of genomic DNA cut with Sst 1 which released the MFG-GC
provirus as a 4.3 Kb fragment which was identified with a cDNA
probe specific for human GC. Western blots and enzymatic assays of
spleen colonies were performed as described above. All five of the
spleens from mice receiving MFG-GC infected bone marrow were
positive for the human GC gene as were all five spleens from mice
receiving the N2-SV-GC infected bone marrow. These results are
summarized in FIG. 6.
Western blot analysis of immunoreactive human GC protein expressed
in spleen colonies of mice sacrificed 12 days after transplantation
with bone marrow infected by the MFG-GC vector was performed.
There was a positive signal about 4-5 times in intensity to the
signal from the human liver control, while the negative control
(normal mouse spleen) showed no signal on Western blot. None of the
colonies from mice receiving N2-SV-GC infected bone marrow showed
detectable expression by Western blot analysis. Enzymatic analyses
of spleen fragments from three mice (A1, A2, and A3) receiving
marrow infected by the MFG-GC vector and of a control colony of a
spleen of a mouse transplanted with normal bone marrow revealed as
seen in FIG. 7 that the GC activity in the spleen fragments was 4-5
fold higher than control spleen background.
In the following example the ability of the MFG-GC vector to
transfer and express the GC gene in transplanted bone marrow was
evaluated.
EXAMPLE 5
EXPRESSION OF GC GENE IN PROGENY OF TRANSPLANTED BONE MARROW
Bone Marrow Transplantation
The ability of the MFG-GC vector to infect and express the GC gene
in the progeny of transplanted murine bone marrow was determined
using the protocol of Bodine, D. M., et al., Exp. Hematol. 19:206
(1991).
Bone marrow cells were harvested from the limbs of
C57BL/6J-Gpi-l.sup.a Hbb.sup.d (HW-80) female mice (Jackson
Laboratories) 3 days after injection with 5-fluorouracil (5-FU)
(150 mg/kg body weight). Bone marrow cells were precultured for 2
days in Fisher's medium supplemented with 15% fetal calf serum, 2
mM L-glutamine, 100 U/ml penicillin, 100 .mu.g/ml streptomycin, and
cytokines. Stem cell factor was provided by Dr. Chris Zsebo (Amgen,
Calif.) and was used in both preculture and coculture procedures.
Initially, 5% conditioned medium from WEHI-3B cells and 5%
conditioned medium from poke weed mitogen stimulated spleen cells
was used as the source of cytokines. Preculture of the bone marrow
cells was followed by 2 days of co-culture with 20 Gy irradiated
viral producer cells (psi-cre) in DMEM supplemented with 10% calf
serum, 100 U/ml penicillin, 100 .mu.g/ml streptomycin, polybrene (8
mg/ml), and cytokines. In later experiments, recombinant IL-3
(Genzyme, Cambridge, Mass.) and IL-6 (gift of Dr. Zsebo) was used
instead of conditioned medium. After coculture, lethally irradiated
mice (Gpi-l.sup.b) were injected with 2.times.10.sup.6 bone marrow
cells for long term hematopoietic reconstitution studies or
1-2.times.10.sup.5 cells for primary day 12 CFU-S assays. Mice were
irradiated with 9.5 Gy of radiation. For secondary recipient
studies, BM from the primary recipient was collected and 10.sup.7
cells were injected into lethally irradiated recipients for long
term reconstitution studies. For secondary CFU-S studies, 10.sup.6
cells were transfused into the recipients.
Transplanted animals were maintained to evaluate the infection and
expression of the GC gene in cells present in the bone marrow
capable of long term reconstitution of the bone marrow of the
recipients. The success of engraftment of transplanted bone marrow
was monitored by the differences in the GPI isoenzymes of the donor
and recipient animals, as described below.
Glucose Phosphoisomerase I (GPI) Isozyme Assay
Hematopoietic reconstitution of the recipient mice, i.e., the
extent of engraftment and reconstitution by donor bone marrow (BM)
cells of irradiated recipient mice, was monitored by the difference
in the electrophoretic mobility of the GPI isozyme (GPI-l.sup.a and
GPI-l.sup.b) present in the donor and recipient leukocytes of the
mice. Such markers permit the estimation of the success of
engraftment of the donor marrow.
GPI isozymes in peripheral blood leukocytes were separated and
analyzed using cellulose acetate electro-phoresis followed by
enzymatic activity staining according to the methods of Eppig, J.
J., et al., Nature 269:517-520 (1977). Donor mice were female
(HW-80) mice (Jackson Laboratories) and the recipient mice were
C57BL/6J-Gpi-l.sup.b (Jackson Laboratories). The results of the GPI
isozyme assay indicated that all recipient animals in this study
had >90% donor cells.
To determine if the engrafted cells were transduced and expressed
human GC in their progeny, estimations were made of peripheral
blood cells using an immunoperioxidase stain specific for the human
gene product using the Vecta stain ABC kit (Vector, Burlingame,
Calif.) according to the manufacturer's instructions. Purified BE4
was used as the primary antibody. The same method was used to stain
cultured macrophages as described below. Cytospin preparations of
peripheral blood were fixed and incubated with the human GC
specific MAb 8E4. The results of these studies on animals at 5
months after bone marrow transplantation are shown in FIGS. 8-11.
The normal BM in FIG. 8 has no staining, whereas in FIGS. 9 and 10
(higher power of FIG. 9) the peripheral leukocytes of a mouse
reconstituted with MFG-GC infected marrow five months previously
showed the presence of human GC enzyme protein as red color.
Approximately 80% of the white blood cells were positive and
several different cell lineages appeared to be expressing the human
gene.
PCR analysis for the human gene was performed on the bone marrow
(BM) and peripheral blood (PB) of three long term reconstituted
mice. The analysis revealed a 192 bp product from the MFG-GC
provirus from the bone marrow and peripheral blood of each long
term reconstituted mouse.
EXAMPLE 6
Long Term Reconstitution Studies
Comparisons of the two retroviral vectors N2-SV-GC and MFG-GC were
performed based on Southern and Western blots and enzymatic
activities in 3T3 cells, spleen foci (CFU-S), LTBMC, and tissues of
mice surviving for longer than 4 months after reconstitution with
bone marrow infected by co-cultivation with the N.sub.2 -SV-GC and
MFG-GC producer lines. Tissues analyzed by Southern and Western
analysis and by enzymatic activity included bone marrow, spleen,
liver, lung, thymus, and lymph node. Protocols of the Southern and
Western analysis and enzymatic assay are described above, as well
as the protocol for preparing and reconstituting bone marrow for
long term recipients.
Eleven HW 80 mice were used in the studies-nine for study of MFG-GC
and two for N2-SV-GC study. The data are summarized in FIG. 11. The
enzymatic activities of the tissues of four mice reconstituted with
MFG-GC infected bone marrow are shown in FIG. 12. The hematopoietic
tissues (spleen, bone marrow, thymus, and lymph node) from those
animals showed GC activities that were in the same range as spleen
colonies and consistently several fold higher than the activity of
control tissues. The hematopoietic tissues reconstituted by
N2-SV-GC infected bone marrow showed little or no increase above
the control activities. These results are consistent with the
activity data on spleen colonies shown in FIG. 6. From the Southern
analysis of hematopoietic tissues it can be seen that both vectors
are equally efficient in infecting early progenitors capable of
long term reconstitution of bone marrow. The copy number was
approximately equal, ranging from about 1 to 2 regardless of the
gene transfer vector used. The data indicate that the efficiency of
transduction approached 100% in cells in bone marrow capable of
long term reconstitution.
In addition, the size of the vector recovered from the tissues was
uniform for each tissue analyzed and was consistent with the vector
used. Restriction with Sst-1 cleaved in the retroviral LTR produced
a 4.9 kb fragment from the N2-SV-GC vector and a 4.3 kb fragment
from the MFG-GC vector. Thus there appeared to be no large
rearrangement of the vector in the long term reconstituted
animals.
As indicated above, however, it is clear that the MFG-GC vector is
transcriptionally more efficient than the N2-SV-GC. Further, the
MFG-GC infected marrow produced cells on a long term basis which
comprised the majority of bone marrow derived cells in the spleen,
bone marrow, and thymus as well as approximately 1 in 10 cells in
the liver.
The data shown in FIGS. 11 and 12 accumulated on non-hematopoietic
tissues (liver, lungs) is also informative. These tissues normally
receive bone-marrow derived cellular elements on a continuing
basis. Under normal physiologic circumstances, bone marrow derived
cells in tissue are primarily macrophages and reflect the normal
provision of tissue macrophages by the bone marrow. In liver,
tissue macrophages (Kupffer cells) constitute approximately 15% of
the cells present at any time. See, Barranger, J. A., N. Eng. J.
Med. 311:101 (1984). If all of the liver macrophages were replaced
in the animals by the progeny of transduced early progenitors in
the bone marrow, the copy number of the vector in the liver should
be about 0.15/cell in long term reconstituted animals. The results
herein demonstrate that both the N2-SV-GC and MFG-GC vectors result
in a copy number in liver of approximately 0.1/cell. This is
consistent with a transduction efficiency approaching 100% as shown
from the data derived from hematopoietic tissues. Furthermore, the
enzymatic activity of the lung and liver of MFG-GC transduced
animals is several fold above background, but is on average less
than that of hematopoietic tissues. This is indicative of the
lesser number of bone-marrow derived cells present in those tissues
and is consistent with the higher enzymatic activities measured in
organs that have a higher proportion of bone marrow derived cells
(e.g., bone marrow, spleen, and lymph node).
EXAMPLE 7
Transduction and Expression of GC Gene in Macrorhages
Macrophages were cultured from the bone marrow of animals whose
bone marrow were reconstituted long term (>4 months) with bone
marrow -transduced by either the N2-SV-GC or MFG-GC vector.
The bone marrows were cultured by the method reported by Gregory,
S. H., et al., J. Leukocyte Biol. 43:67 (1988). Conditioned media
containing macrophage colony stimulating factor (M-CSF, also
designated as CSF-1) was used to culture the whole bone marrow from
mice reconstituted with MFG-GC infected bone marrow and surviving
longer than five months. 2.times.10.sup.5 bone marrow cells from
the long term reconstituted mice were suspended in 20 ml of DMEM
supplemented with 4.5 g/l glucose, 3.7 g/l NaHCO.sub.3 10% heat
inactivated FBS, 20% heat inactivated horse serum, 20% L-929 cell@
conditioned medium (gift of Dr. Sallie Boggs), 100 U/ml penicillin,
100 .mu.g/ml streptomycin, 200 .mu.g/ml glutamine and then
cultivated in 100 mm dishes at 37.degree. C. and 5% CO.sub.2. The
L-929 conditioned medium contained mouse macrophage colony
stimulating factor (M-CSF). See, Stanley, E. R., et al., J. Biol.
Chem. 252:4305 (1977). Such a culture system resulted in
essentially pure colonies of macrophages.
Bone marrow cultured with this medium produced pure macrophage
cultures that expanded approximately five logs. After 11 days,
latex beads were added to the medium to confirm that the adherent
cells were of macrophage lineage. The cells were harvested and
assayed by measurement of GC enzymatic activity,
immunocytochemistry, Western blot, and Southern blot as described
previously. The results are described below.
Cultures of macrophages established from MFG-GC, control, and
N2-SV-GC animals were immunohistochemically stained as described in
Example 5 for the human GC gene product. FIG. 13 shows the culture
from an MFG-GC reconstituted animal which showed the ability of the
cultured macrophages to phagocytose latex beads. This same culture
was very positive for expression of the human GC gene product since
virtually every cell was stained as seen in FIG. 14. Cultures from
a control animal, FIG. 15, and N2-SV-GC reconstituted animal, FIG.
16, were negative for the expressed human gene product. Thus the GC
gene was not expressed in macrophages when starting bone marrow was
infected by N2-SV-GC.
MFG-GC macrophages cultured from explanted bone marrow obtained
from eight different mice reconstituted long term with MFG-GC
infected bone marrow and macrophages cultured from a control mouse
were assayed for GC enzymatic activity. As seen in FIG. 17A, the
enzymatic activity of the MFG-GC macrophages (lanes 1-8) was on
average approximately 5 times that of the control macrophage
activity (lane 9). Western blot analysis in FIG. 17B of
immuno-reactive human GC protein expressed in extracts from
macrophages cultured from explanted marrow of a long term
reconstituted mouse rescued with marrow infected with MFG-GC (lane
1) and in extracts from a control mouse macrophage culture (lane 2)
showed that the protein of the expected size range (59-66 Kd) is
expressed by the MFG-GC transduced macrophages (lane 1). In FIG.
17C Southern blot analysis of DNA from macrophages cultured from
explanted marrow of a long term reconstituted mouse rescued with
marrow infected with MFG-GC (lane 1); a normal mouse control (lane
2); and standards for estimating copy number representing 0.2, 0.5,
and 1.0 copies/cell, respectively (lane 3) showed the expected 4.3
Kb band for the DNA from the long term reconstituted animal (lane
1). The DNAs were probed with human .sup.32 P-GC-cDNA. The copy
number in these cells was approximately 1/cell.
Such results confirm that the transduction efficiency of the MFG-GC
and N2-SV-GC-vectors was very high for early progenitors,
approaching 100%. However, as seen in the macrophage studies,
MFG-GC is a superior vector for the expression of the GC gene in
bone marrow derived macrophages.
Secondary Transplants
As a further measure of the ability of the MFG-GC vector to
transduce early progenitors, secondary bone marrow transplantations
using bone marrow from long term reconstituted MFG-GC mice were
performed. FIG. 18A shows Southern blots of 27 spleen colonies from
three mice isolated 12 days post-transplant. DNA was digested with
Sst I and probed with full length human GC cDNA. Lanes 1-10
contained DNA from spleen colonies isolated from mouse #26; lanes
11-19 contained DNA from spleen colonies isolated from mouse #27;
lanes 20-27 contained DNA from spleen colonies isolated from mouse
#28; lane 28 contained genomic DNA isolated from an age-related
control; and lanes 29-31 contained standards for estimating copy
number (0.2, 0.5, and 1.0 copies/cell, respectively). All
twenty-eight foci (28/28) were positive for the human gene.
Enzymatic analyses of the same 27 spleen colonies are shown in FIG.
18B. Lane 1 represents GC activity of an lanes 2-10, 11-19, and
20-27 represent, respectively, spleen colonies from mouse #26, #27,
and #28, showing GC activity approximately 2-6 times that of the
control, thereby providing more evidence that the MFG-GC vector was
able to efficiently transduce stem cells.
EXAMPLE 8
Transduction and expression of the Human GC Gene in Normal Human
and Gaucher Patient Macrophages (MO)
Human macrophages were cultured from peripheral blood from normal
volunteers and patients with Gaucher disease following an approved
protocol (IRB # 910505). Blood was collected in 50 cc syringes
containing 2.5 ml of Na heparin (1000.mu./ml). 6% Dextran was then
added in saline solution in an amount that was approximately 10% of
the blood volume (to a 50 ml syringe, 5 ml of Dextran was added).
The resulting composition was then mixed thoroughly and left
undisturbed at room temperature for 40 minutes until a clean
separation line was visible between the supernatant and the red
blood cell (RBC) layer. Using a bent 18G needle, the supernatant
was layered onto Ficoll-Hypaque (Histopague.RTM. 1077, Sigma)
(F-H). The tubes were then centrifuged for 20 minutes at 530 g
(1300-1500 RPM) at room temperature. A mononuclear cell (MNC) layer
was easily visible as a white band in the middle of the
supernatant. The superantant above the MNC layer was aspirated and
disposed. The MNC layer was transferred to a separate tube for
washing the Ficoll from the cells. Approximately 20-40 ml of
Dulbecco's phosphate buffered saline (D-PBS) (Gibco) (1.times.) was
added to the tubes containing the MNC's, or alternatively,
KRP-glucose could be used. The cells were washed thoroughly at
least two times, and the pellets were resuspended in culture medium
(1640 RPMI (Gibco), 10% FBS, 10% human albumin). The cells were
then counted and a viability assay was performed using trypan blue
exclusion. Cytospin slides for Wright staining were prepared and
the percentages of resultant isolated macrophages and lymphocytes
were counted.
In order to culture macrophages, the macrophages from the preceding
step were used. The cell count was adjusted to .apprxeq.5.times.106
cells/ml with culture medium (1640 RPMI, 10% FBS, 10% human
albumin), and the cells were plated in 35 mm culture dishes
(approximately 10 cm.sup.2 surface) so that the plating density was
between 0.2-0.6.times.10.sup.6 macrophages/cm.sup.2. The cells were
then incubated at 37.degree. C. in 5% CO.sub.2 for 60 minutes, and
non-adherent cells were removed after the incubation. Then 1 ml of
D-PBS (1.times.) was gently placed in the dish, and dish was shaken
to remove non-adherent cells left in the bottom of the dish. 2 ml
of previously warmed culture media was then added to each 35 mm
dish, and the cells were then incubated at 37.degree. C. in 5%
CO.sub.2. The culture medium should be changed only once a
week.
In order to infect human macrophages in culture with MFG-GC, the
culture medium from the cultured macrophages of the previous step
was removed from each culture dish and 1 ml of viral superantant
obtained from, the producer cell line described above was added to
each culture dish, along with 10 .mu.l of Polybrene (8 .mu.g/ml).
The dishes were then incubated at 37.degree. C. in 5% CO.sub.2 for
48-72 hours. The plates were then checked to assure an adequate
number adherent cells were present. The culture medium containing
the viral supernatant was then removed from each culture dish, and
each plate was washed with D-PBS (1.times.), and the D-PBS was
removed. Then, 1 ml of Lysis buffer (0.05M K-P, 1% Triton X-100, pH
6.5) was added to each plate, and a rubber policeman was used to
detach all cells from the bottoms of each dish. The cell lysates
were then transferred to small Eppendorf tubes and placed on ice.
The lysates were sonicated and processed for GC specific enzymatic
activity and protein assays as described previously.
As shown in FIG. 19 the enzymatic activity assays of these cells
demonstrated that the MFG-GC vector is able to impart approximately
75 U/mg cell protein to either normal macrophages (A) or the
macrophages from patients (B). This increment of enzyme is
therapeutically effective to correct the enzymatic deficiency
completely in the Gaucher disease patients. The present invention
is therefore useful as a method of treating Gaucher disease in
human Gaucher patients.
In the following example the absence of replication competent
helper virus from viral stocks and tissues of mice transduced with
MFG-GC was shown.
EXAMPLE 9
Mobilization Assay for Helper Virus
Supernatants from psi-cre ecotropic producer lines, the PA 317
amphotropic producer lines, and from bone marrow and homogenates of
spleen cells from the long term reconstituted animals were assayed
by the BAG mobilization method assay for the helper virus. The
assay was carried out essentially according to the methods
described by Danos, O. Methods in Molecular Biology, Vol. 8:
Practical Molecular Virology: Viral Vectors for Gene Expression,
Ed. Collins, M. (Humana Press, Clifton, N.J.) 17-26 (1991). The
indicator line (3T3-BAG) was prepared by transducing NIH 3T3 cells
with the BAG virus, which contains both the .beta.-galactosidase
and the neo R genes. The 3T3 BAG cells (2.times.10.sup.5) were
cocultured with recipient mouse bone marrow cells
(5.times.10.sup.6), spleen cells, and superantants from the viral
producer lines in 10 cm tissue culture dishes in the presence of 8
.mu.g1m1 polybrene. When the cells reached confluence, the medium
was removed and 4 ml of fresh medium was added to each dish. The
supernatant was harvested after 16 hours, filtered through a 0.45
micron membrane filter, and stored at -80.degree. C. until the
assay was performed. The cycle of splitting and harvesting
supernatant was repeated several times. Supernatants were analyzed
for the presence of helper virus by infecting 3T3 cells with 2
ml/10 cm dish or 1 ml/6 cm dish for 2 hours at 37.degree. C. in the
presence of 8 .mu./ml polybrene. Fresh medium was then added, and
after 48 hours, the cells were harvested and assayed for lacZ
expression by Xgal staining and for G418 resistance by replating
5.times.10.sup.5 cells in the presence of G418 (400.mu./ml).
Control infections of 3T3 cells were performed with serial
dilutions of BAG viral supernatants. These cells were positive for
lacz expression and G418 resistance.
Supernatants from the psi-cre #4 and the PA-317 virus producing
line (VPL) did not result in the generation of any blue cells in
the BAG mobilization assay nor did any cells grow through G418
selection. Minced spleen and bone marrow from long term
reconstituted mice were also studied by culture on BAG cells and
compared to control tissues. In one of two controls, minced spleen
co-culture resulted in one blue 3T3 cell in 2000 cells. The cells
were further grown in G418 containing media. The supernatants from
these G418 resistant cells were used to infect fresh 3T3 cell
cultures. Again, 1 blue cell in 2000 cells was seen. No blue cells
were seen in BAG cell controls. These results indicate the BAG
mobilization assay detects low levels of a virus in control mouse
tissues that is capable of rescuing the BAG-neo gene, but is unable
to replicate or efficiently carry on cycles of infection. Similar
results were seen in one of four spleen cocultures from long term
reconstituted mice. These results demonstrate that the viral stocks
and recipient animals are free of replication competent helper
virus. Absence of a helper virus was confirmed in the PA-317 VPL by
an independent laboratory (compliments of Dr. Frederick
Schuening).
EXAMPLE 10
Enrichment of Human CD34+ Cells
Hematopoietic cells were applied to a Ficoll-hypaque gradient, and
the light density cells were washed and resuspended at
1-2.times.10.sup.8 /ml in phosphate buffered saline (PBS)
containing 1% bovine serum albumin (BSA). Biotinylated anti-human
CD34+ monoclonal antibody (80ul per ml) was added and cells were
incubated at room temperature for 25 minutes. The antibody-labeled
cells were applied to the prepared avidin column, followed by a
wash with PBS containing 1% BSA. The biotinylated anti-CD34+
antibody, the avidin column, and necessary reagents (CEPRATE LC
(CD34+) Cell Separation System) are supplied by CellPro, Inc.,
Bothell, Wash. Adsorbed CD34+ cells were released by mechanical
manipulation.
The results of a typical >10 fold enrichment, with 87.6% purity,
are shown in FIG. 20. In these laboratory scale columns, enrichment
is less than in the large clinical scale columns. Increased numbers
of colony forming units (CFU) were consistently noted with CD34+
cells (B), compared to the pre-enriched samples (A). Clonogenic
assays revealed an increase in CFU of about 3 fold (FIG. 21).
Prestimulation and Transduction of Human CD34+ Cells
Human long-term bone marrow culture medium (LTCM) was used for
prestimulation and expansion of human CD34+ cells and for long term
cultures. It consists of Iscove-Modified Dulbecco's Medium (IMDM)
(Gibco BRL) containing 12.5% fetal bovine serum (FBS) (Hyclone
Laboratories), 12.5% horse serum (HS) (Hyclone Laboratories), 2 mM
L-glutamine, 1.times.10.sup.-6 M 2-mercaptoethanol,
1.times.10.sup.-6 M alpha-thioglycerol, 1 .mu.g/ml hydrocortisone,
and penicillin/streptomycin. CD34+ PB cells were exposed to viral
supernatants 6 times over a period of 5 days, beginning three days
after incubation, in mixtures of cytokines, as indicated. Most of
the cells were harvested one day after the final infection and were
analyzed for enzyme activity. Values given are the means for
duplicate pellets. Results of experiments with human CD34+ enriched
PB cells from G-CSF primed lymphoma patients show that transduction
occurs and leads to expression of the glucocerebrosidase gene in
these cells and their progeny using the MFG-GC retroviral vector.
FIG. 22 shows enzyme activities of cells harvested shortly after
infection in three different mixtures of cytokines: 1) IL-3, IL-6,
SCF, and GM-CSF; 2) IL-3, IL-6, and SCF; and 3) SCF and PIXY. In
each case, the enzyme activity in the MFG-GC infected group was
more than 80% higher than the level of endogenous activity in
noninfected cells grown under the same conditions. Control samples
infected with a retroviral vector, containing the bacterial lacZ
gene, did not differ significantly from noninfected cells,
eliminating the possibility that increased activity from MFG-GC
transduced cells was a result of nonspecific viral exposure or
infection conditions. Samples of infected and control CD34+ cells
were harvested and analyzed for GC enzyme activity at the times
indicated, while the remaining cells were further expanded in
culture over a period of three weeks following infection. Increased
enzyme expression from the MFG-GC vector was evident immediately
after infection, and this enhancement did not diminish with
continued time in culture. Infected cells were expanded in vitro
for three weeks following infection, and a sustained elevation was
observed in enzyme activity in the experimental group versus
control cells (FIG. 23). Southern blot hybridization of DNA from
these cells suggested a transduction efficiency of 5 to 15%.
In two subsequent experiments (Experiments #2 and #3),
significantly increased GC expression was not observed following
infection procedures. In these experiments, rat recombinant SCF
(rSCF) was used during infection and expansion of the cells.
Although the biological activity of this factor was expected to be
similar to that of human recombinant SCF (hrSCF) (Langley, K. E.,
et al., Arch. Biochem. Biophys. 295(1):21-28, 1992), it became
clear in Experiment #4 that the cytokine activity was suboptimal
using rSCF from this particular source. Following infection and
expansion into medium containing rrSCF, cell growth was observed to
diminish. Aliquots of two samples were transferred to two dishes
each, and to one dish from each sample was added hrSCF. These
samples were incubated for an additional 6 days prior to counting
cell numbers in a hemacytometer (FIG. 24).
In the next experiment (#4), cells were harvested at the point at
which a decline in growth rate was observed. Analysis of
glucocerebrosidase activity in these cells indicated that elevation
of activity had occurred in each of several different groups
following infection, in comparison to similarly cultured
noninfected cells. These cells had been infected under a variety of
conditions in an effort to uncover a reason for the lack of
response in the two previous experiments. Two mixtures of cytokines
were used: 1) IL-3, IL-6 and SCF, or 2) PIXY, IL-6 and SCF. The
latter group was further subdivided into cells infected in the
presence of a stromal cell layer, in the presence of human serum,
on Petri dishes and on treated tissue culture dishes. The specific
activity of these samples was lower across all the groups than in
previous experiments (control cells .about.150 U/mg vs >200
U/mg), perhaps as a result of the declining viability of these
cells at the time, but the elevation for infected cells was about
70% above endogenous levels (FIG. 25).
In Experiment #5, for which recombinant human SCF was used
throughout infection and expansion phases, the specific activity
was once again >200 U/mg, an increase of more than 100% above
endogenous GC activity. FIG. 26 provides a summary of experiments
1-5 showing GC enzyme activity in human CD34+ PB cells following
infection with MFG-GC containing supernatants.
Southern blot hybridization of Sst 1 digests of DNA from infected
and noninfected CD34+ enriched PB cells was performed (Experiment
5). Copy number controls were prepared using known amounts of DNA
from a murine monoclonal cell line containing 5 copies per cell.
Sst 1 cuts within the LTRs on each end of the vector to release the
4.3 kb fragment (apparent mol. wt.), to which the 1.8 kb GC cDNA
probe hybridizes. In human cells the probe also binds strongly to
the 6.4 kb Sst 1 fragment of the normal GC gene and the pseudogene.
Southern blot hybridization suggested a transduction efficiency of
>20% (FIG. 27).
FIG. 28 shows the results of subsequent experiments, which
demonstrate that normal CD34+ cells are transduced and express 2-4
times the normal activity of GC. Some of the variables that affect
the transduction efficiency and resultant expression of GC in human
CD34+ cells have also been explored. It is believed that the
combination of IL-3, IL-6 and SCF (each at 10 ng/ml) provides the
most consistent results for GC expression.
Human bone marrow CD34+ cells were infected by daily exposure to
virus-containing supernatant for 4 days. Prior to the second and
each subsequent infection, an aliquot of cells was transferred to a
separate dish without further exposure to virus. After the
infection period, cells were expanded and harvested for-analysis of
enzyme expression. There is a linear relationship between total
number of doses of viral supernatants and GC activity expressed in
CD34+ cells (FIG. 29).
Using the conditions described above, CD34+ cells obtained from the
bone marrow of Gaucher patients have shown an increase in GC
activity of 20-40 fold, essentially equaling the results in
transduced CD34+ cells from controls (FIG. 30). These cells kept in
culture for 3 weeks maintain the elevated levels of
glucocerebrosidase activity.
Gaucher CD34+ cells were rapidly evaluated for transduction
efficiency using a conventional chromogenic immunocytochemical
assay for glucocerebrosidase. Immunocytochemical staining for human
glucocerebrosidase (GC) utilized cytospin preparations stained with
the monoclonal antibody 8E4. Using this procedure, as shown in FIG.
31, CD34+ cells from a Gaucher patient gave no detectable color
formation (B), whereas, many cells in the infected population gave
a strong signal (see FIG. 31). Counting ten high power fields and
scoring for immunocytochemical positive cells, an estimated
transduction efficiency of approximately 20% was calculated.
The colony forming efficiency in methylcellulose was determined for
freshly enriched CD34+ cells and for infected and control cells
following infection procedures. Results from many experiments
indicated that between 30-100% of the colony forming efficiency
remains after infection.
Optimal Conditions for Transduction of Human CD34+ Cells
Procedures for the optimal transduction of CD34+ cells have been
determined (Bahnson, A. B., et al., Gene Therapy 1(3):176-184,
1994; Bregni, M., et al., Blood80:1418-1422, 1992; Langley, K. E.,
et al., Arch. Biochem. Biophys. 295(1):21-28, 1992). Centrifugation
of target cells with viral supernatant was examined. Initially, the
results of this procedure was measured in a cell line (TF-1) and in
3T3 cell targets. These studies demonstrated activities of GC in
excess of 10 fold above the baseline in these targets. The
parameters of optimal speed of centrifugation, length of
centrifugation, shape of vessel, number of procedures, and
temperature at which the procedure is performed was examined. The
conclusions from these experiments were that a speed of
10,000.times.g in a round bottom tube repeated 2-3 times over
several days, was optimal. The length of time in the centrifuge
increased the resultant GC activity in the targets and was linear
out to 400 minutes. Room temperature or 37.degree. C. did not make
a significant difference. Centrifugation was then applied to CD34+
cells to examine its value on the transduction of these cells for
use in clinical gene therapy protocols for Gaucher patients.
A protocol was used in which 5.times.10.sup.5 CD34+ cells were
centrifuged at 2400.times.g for 2 hours with either 0.2 ml or 1.0
ml of viral containing supernatant with a titer of 5.times.10.sup.6
pfu/ml. The results of these studies showed that three infections
were optimal and yielded enzymatic GC activity of 10-15 times that
of uninfected CD34+ cells. PCR analysis of CFU-GM derived from
these cells revealed a transduction efficiency of 100%. The GC
activity of CD34+ cells expanded for 17 days in culture remained at
levels >5 times the control, uninfected cells.
EXAMPLE 11
Construction of R-GC Vector
The R-GC retroviral vector was developed from the MFG-GC vector, by
the insertion of a Sac II linker placing the gag sequences out of
frame thus rendering the vector unable to synthesize truncated
gag-related peptides. It also provided another site of sequence
dissimilarity making recombination events to wild type less of a
possibility. After partial Sma I digestion of the vector, this
additional safety step was accomplished by inserting an eight base
pair Sac II linker 3' to Hae III site at the ATG of the half-gag
gene in the retroviral sequence.
The retroviral vector, R-GC, containing the gene for
glucocerebrosidase (GC) is a replication defective vector. The
structure of R-GC is shown in FIG. 32.
Generation of the YCRIP GC Producer
The viral producer line of R-GC was developed by infection of the
psi crip packaging line (Danos, O., et al., Proc. Natl. Acad. Sci.
USA 85:6460-6464, 1988) by supernatant from BOSC cells (Pear, W.
S., et al., Proc. Natl. Acad. Sci. USA 90:8392-8396, 1993)
transfected by the R-GC plasmid. Selection of high titer producer
cells was done by GC enzymatic assay screening and confirmed by
Southern blot analysis. The producer is free of bacteria,
mycoplasm, and replication competent retrovirus.
The producer line was obtained using a two step procedure (FIG.
33). First, the plasmid form of the vector was transiently
transfected using calcium phosphate/DNA precipitation into BOSC 23
cells (Pear, W. S., et al., Proc. Natl. Acad. Sci. USA
90:8392-8396, 1993), from which were obtained ecotropic
vector-containing supernatants. Secondly, these supernatants were
used to cross-infect amphotropic YCRIP packaging cells. High GC
enzyme levels (.about.10.times. control) in the infected YCRIP
cells indicated that the BOSC 23 cells had efficiently packaged the
R-GC vector. Clones were obtained from the infected YCRIP cells
using limiting dilution without selection. Supernatants from these
clones were screened by infection of 3T3 targets (FIG. 33), and the
highest titer clones were selected for further
characterization.
Gene Transfer and Expression in 3T3 Target Cells Using the R-GC
Vector
Following the initial screening, the cells were expanded into
monolayers for comparison of the highest titer clones under similar
conditions of confluence. With the amphotropic producer of MFG-GC
(cc-2), maximal titers are obtained with fully confluent and/or
"overconfluent" monolayers (Bahnson, A. D., et al., Gene Therapy,
1(3):176-184, 1994), and similar results have been obtained for the
highest titer amphotropic producer of R-GC (LM30) (FIG. 34). The
four highest titer clones were compared under conditions of
subconfluence and "overconfluence" for production of
vector-containing supernatants. Supernatants were assayed on 3T3
targets using standard procedures.
Based on the above and additional data, clone LM30 was chosen for
expansion and further characterization. To test the stability of
these cells, cryogenically preserved LM30 cells, were thawed and
re-expanded to confluent monolayers, from which supernatants have
been tested and compared with previously prepared supernatant.
The ability of the LM30 clones to generate a titer that is
approximately equal to the original titer was tested over a period
of two years. The original clone was produced in 8/92. Frozen vials
prepared in 6/93 were thawed in early 1995 and supernatants
generated from the producer over several weeks. These supernatants
were able to impart enzymatic activity on target cells that was
equal to or greater than the activity conferred on the target cell
line more than two years earlier (FIG. 35).
These cells have been used to generate supernatant for use in a
variety of tests, including the verification of equivalence of the
R-GC vector to the MFG-GC vector, the ability to infect and yield
GC expression in human CD34+ cells, demonstration of stability of
the producer and of the vector-containing supernatants over time,
and additional investigations into possibilities for concentration
of virus and optimization of conditions for efficient
transduction.
EXAMPLE 12
R-GC Gene Transfer and Expression in Human CD34+ Cells From
Blood
The GC enzyme activity of R-GC infected CD34+ cells from the
peripheral blood of two Gaucher patients was more than two fold
above control noninfected cells after 1.times. infection and about
four fold above controls after 3.times. infections with supernatant
from the LM30 clone (FIG. 36). In these experiments, supernatants
from the cc-2 clone (producer of nonmodified MFG-GC) were also
tested for comparison. R-GC and MFG-GC were obtained from the YCRIP
producer cell lines, LM30 and cc-2, respectively. The cc-2
supernatant was from the highest-titer pooled batch,
P3.times.8--Aug. 24, 1993. The LM30 supes were from the
highest-titer individual supernatants available at the time:
PB9=LM30 3.times.6--Dec. 20, 1993, and PB10=LM30b 3.times.9--Jan.
7, 1994. The cells from PB9 were 5% CD34+ by FACS analysis, and
were infected with only one addition of supernatant. The cells from
PB10 were more highly enriched for human CD34+ (.about.40%), and
were infected with three exposures to fresh supernatant over 1.5
days. Both protocols included a 24 hour prestimulation in medium
containing IL-3, IL-6, and SCF. Results for enzyme activity are
expressed relative to mock-infected control cells exposed to
10%CS/DMEM and cultured in an identical manner to the infected
cells in each experiment. Noninfected control activities were 165
and 44 U/mg for PB9 and PB10, respectively.
The R-GC vector was used to infect 8 separate samples of CD34+
cells obtained from two patients. Infection was carried out in
sterile 150 ml plasma transport bags using a single centrifugation
promoted infection protocol. GC activity was restored to at least
normal in each experiment (FIG. 37).
FIG. 38 shows the GC activity of R-GC transduced CD34+ cells from a
single Gaucher patient in multiple studies, in which GC activity
levels exceeded uninfected normal cells and uninfected Gaucher
cells. The mobilized cells were infected by a single infection.
EXAMPLE 13
PRODUCTION OF THE VECTOR SUPERNATANTS AND GENETICALLY CORRECTED
CELLS
Master Cell Bank
A clone from the virus-producing cell line (LM 30.2.7) has been
used to generate a lot-specific Master Cell Bank (MCB) in the Human
Gene Therapy Applications Laboratory at the University of
Pittsburgh. These banks were preserved at a low passage number. To
establish a MCB, virus producer cells were grown in the standard
medium in static culture and expanded in number by serial transfer
to produce sufficient quantity of viable cells to freeze down a
single lot of 100 ampules containing 1.times.10.sup.7 cells
each.
Complete medium for retroviral vector-containing supernatant
production consists of: Dulbecco's Modified Eagle's Medium (DMEM),
high glucose, L-glutamine (2 mM), and 10% calf serum. Also applied
to the cells at the time of splitting for expansion are: Dulbecco's
Phosphate Buffered Saline (PBS; 2.7 mM KCl, 1.2 mM KH.sub.2
PO.sub.4, 138 mM NaCl, 8.1 mM Na.sub.2 HPO.sub.4) and Trypsin-EDTA
(0.05% Trypsin, 0.53 mM EDTA in Hank's Balanced Salt Solution
without Ca and Mg).
Producer Cell Culture Method
LM 30.2.7 cells were thawed from cryopreserved material, washed
briefly in complete medium to remove DMSO, and plated in complete
medium in sterile plastic tissue culture dishes. The dishes were
incubated in Forma Steri-Cult 200 Incubators at 37.degree. C. in an
atmosphere of 5% CO.sub.2 in air with 90% relative humidity. They
were monitored microscopically, and over several days' time the
cells expanded in number until reaching confluence. The cell
monolayers were then washed briefly with PBS, detached from the
plastic surface with trypsin, mixed into a uniform suspension, and
distributed (replated) into new dishes. The number of new dishes
provides 10 to 30 times the surface area of the starting dish(es).
For production of a cell bank, the process of cell number expansion
followed by trypsinization and replating is repeated until
sufficient cells were generated. As is customary, 100 vials
containing 10.sup.7 cells per vial were put up for a cell bank.
Vector Supernatant Production (Lot)
For R-GC supernatant production from LM30.2.7, the expanded cells
from the working cell bank are allowed to reach confluence, and
medium is then replaced on a daily basis. The medium removed is
frozen and stored at -80.degree. C. in sterile containers of
sufficient size to accommodate collection at each time point. At
least one aliquot from each individual collection is frozen in a
smaller container for determination of viral vector titer.
Collection is continued daily for a period of up to two weeks.
The test aliquots are assayed for biologic activity, and acceptable
individual collections are mixed in a single sterile container to
form a uniform lot. This lot is then filtered through a membrane
filter (0.45 .mu.m, sterile non-pyrogenic cartridge type) using a
peristaltic pump with nontoxic, nonpyrogenic tubing, and the
filtered final product is refrozen in sterile containers and stored
at -80.degree. C. until use.
Stability Testing
Stability studies were performed to assess the stability of the
frozen viral supernatant.
The stability of high titer producer, cc-2, from MFG-GC was studied
over a 16 month period. The supernatant was stored at -80.degree.
C. and samples were tested every three months. Supernatant was
originally collected on Aug. 14, 1992 from freshly confluent
monolayers of cc-2 producers in two 15 cm dishes. Volumes of 15 ml
of medium (2.times.) were collected following 7 hours of
conditioning from each dish. This supe was filtered (0.45 um) and
stored in five 15 ml tubes, 5 ml per tube, at -80.degree. C. Tubes
have been thawed and re-aliquoted into 1 ml tubes as necessary over
the period of time shown. For infection, 0.5 ml of supernatant is
mixed with 0.5 ml of medium and 8 .mu./ml of polybrene and applied
to 3.times.10.sup.5 target 3T3 cells on 6 cm dishes
(1.5.times.10.sup.5 cells seeded one day previous). The cells are
incubated at 37.degree. C. and gently rocked every 15 minutes for 2
hours, at which time 2 ml of fresh medium is added to each dish and
incubation is resumed until harvest 2 days later. The data suggest
that the supernatant is stable over the test period of 12 months
(FIG. 39).
To test the stability of R-GC/LM30 clones, cryogenically preserved
LM30 cells were thawed and re-expanded to confluent monolayers,
from which supernatants have been tested and compared with
previously prepared supernatant. The titers of these supernatants
were not significantly different from those obtained prior to
cryogenic storage.
EXAMPLE 14
ISOLATION, ENRICHMENT AND TRANSDUCTION OF AUTOLOGOUS CD34+
CELLS
G-CSF Mobilization
G-CSF mobilization is used in patients with Gaucher Disease to
increase the number of CD34+ cells in the blood. Candidates receive
G-CSF at a dosage of 5 .mu.g/kg/day by subcutaneous injection on
consecutive days.
Leukopheresis
Leukopheresis procedures are continued on a daily schedule until a
total mononuclear cell (MNC) yield of 7.times.10.sup.8 /kg is
obtained. It is anticipated that 3-5 collections are necessary.
Seven to ten (7-10) liters of the patient blood is processed per
procedure, depending on the patient's total blood volume
hematocrit. Each leukopheresis product is enriched for CD34+ cells
on the day of collection.
Enrichment of CD34+ Cells
Leukopheresis product is enriched for CD34+ cells using the
Ceprate.TM. SC Stem Cell Concentrator (CellPro, Inc.). The cells
are labeled with the 12.8 biotinylated anti-CD34+ antibody and
loaded on to a CellPro column which permits separation of the CD34+
cells. Typical enrichment is 40-50 fold and the cells are usually
more than 80% pure by FACS analysis. At the end of the separation,
the CD34+ cells are viably frozen. The CD34+ cells are pooled
before transduction.
Transduction of CD34+ cells
The pooled human CD34+ cells from peripheral blood are suspended in
medium containing a mixture of cytokines. The mixture consists of
rhSCF (10 ng/ml), rhIL-6 (10 ng/ml) and rhIL-3 (10 ng/ml) and long
term bone marrow culture medium (LTBMC). LTBMC consists of Iscove's
Modified Dulbecco's Medium (Gibco BRL) containing 12.5% horse serum
(Hyclone Laboratories), 12.5% fetal bovine serum (Hyclone
Laboratories), 1 mm mercaptoethanol, 1 mm alpha thioglycerol, 2 mM
L-glutamine, and 1 mg/ml hydrocortisone. Cytokines are purchased
from Preprotek, Inc. (Rocky Hill, N.J.). The cells are placed into
a 10 cm dish at an approximate concentration of 2.times.10.sup.5.
The cells are pre-incubated in this mixture for 24 hours.
For infection, the viral supernatant is added to the cells, three
times over 36 hours. Protamine sulfate is added to yield a final
concentration of 4 mg/ml. Cell number is measured by counting
suspended cells using a hemacytometer. The cells are pelleted by
centrifugation at 500 g for 15 minutes. The pellet is then washed
twice in cold PBS and resuspended at a cell concentration of
1.times.10.sup.7 /ml and viably frozen. The frozen cells are thawed
slowly to maintain viability and prepared for administration to the
patient by washing twice in PBS.
Testing of the Transduced CD34+ Cells
Transduction efficiency of the transduced CD34+ final product is
monitored by PCR of CFU-GM clonogenic assays to determine the
presence of the GC gene, Southern blotting to determine GC copy
number, GC enzyme activity, or immunocytochemistry to detect GC
protein.
For PCR analysis, a reaction mixture is made and aliquotted to each
of the sample tubes, such that the final PCR reaction contains 200
.mu.M of each dNTP, 0.5 units Taq polymerase (Amplitaq.RTM.
(Perkin-Elmer)), 2 mM MgCl.sub.2, and 0.4 .mu.M of each primer in
dH.sub.2 O. The final reaction volume is 50 .mu.l. A pair of
primers AB1 and AB2 is used, one which hybridizes to the GC cDNA
region and the other to the viral sequence, respectively, to yield
a unique 407 bp amplification product. (primers AB1: 5' ACG GCA TGG
CAG CTT GGA TA 3' (SEQ ID NO: 1), AB2: 5' AGT AGC AAA TTT TGG GCA
GG 3' (SEQ ID NO: 2)). Thermal cycling is performed on a Gene Amp
PCR system 960 as follows: 94.degree. C..times.5 minutes for an
initial denaturing cycle, then 30 cycles of 94.degree. C..times.45
seconds, 58.degree. C..times.45 seconds, 72.degree. C..times.30
seconds. The PCR products are resolved on a 6% acrylamide or 2%
agarose gel. The bands are visualized by ethidium bromide and UV
light.
Cells to be assayed for GC activity are washed twice in PBS,
pelleted and stored at -80.degree. C. prior to analysis. Pellets
are thawed and extracted in cold 50 mM potassium phosphate buffer
(pH 6.5) containing 2.5 mg/ml Triton X-100. Ultrasonification is
used to disperse and lyse the cells, followed by 15 minute
centrifugation in a microfuge at 4.degree. C. to yield a clear
supernatant for analysis Enzyme activity is determined by addition
of 10 mM 4-methylumbelliferyl-.beta.-D-glucopyranoside (Sigma
Chemical Co.) in citric acid-sodium phosphate (0.12M) buffer (pH
5.4) containing 2.5 mg/ml sodium taurocholate, 2 mg/ml Triton
X-100, and 10 mg/ml bovine serum albumin. The reaction is
terminated after 30 minutes at 37.degree. C. by addition of 0.17M
glycine-carbonate buffer (pH 10.4), and the fluorescence of the
4-methylumbelliferone product is measured with a fluorometer.
Protein concentration is determined using bicinchoninic acid
according to manufacturer's instructions (Pierce Chemical Co.). The
specific enzymatic activity of GC is expressed as nmoles per hour
per mg of protein (U/mg).
Immunochemical detection of the human GC protein is performed
according to manufacturers directions using the Vectastain ABC Kit
(vector Laboratories). This kit contains a biotinylated antimouse
IgG and produced an avidin biotin peroxidase complex monoclonal
antibody 8E4 was used as the primary antibody. Cells are identified
by staining with the horseradish peroxidase substrate
3-amino-9-ethyl-carbazole (Sigma Chemical Co.) and by
counterstaining with hematoxylin. Negative controls are Gaucher
patient cells. Positive controls are the CC-2 cell line. These
producer cells are known transduced cells expressing the human GC
protein.
EXAMPLE 15
GENE THERAPY FOR GAUCHER DISEASE
The CD34+ enriched fraction of autologous cells isolated from a
Gaucher patient and transduced with R-GC to genetically correct the
inability to synthesize functional GC enzymatic activity (verified
as described in Example 14, supra) is transplanted back to the
patient, without bone marrow ablation. A therapeutically effective
number of the transduced CD34+ cells (e.g., 2.times.10.sup.6 /kg)
are reintroduced into the patient by infusion, so as to provide a
therapeutically effective level of GC activity.
Periodically thereafter, a sample of blood from the patient is
removed and peripheral blood leukocytes (PBL) are isolated to assay
for GC enzyme activity. If the GC activity is not increased at
least 2-fold above the pre-treatment level, transplantation is
repeated. Transplantation may be repeated up to four times in the
first year.
PBL are monitored monthly to assay for the presence of the
transduced GC gene, and for GC enzyme activity (as described in
Example 14, supra).
Clinical responses to the gene therapy are monitored in treated
Gaucher patients by routine blood profiling including hemoglobin
level and platelet count, size of liver and spleen, bone marrow
biopsy, and x-rays of long bones and the pelvis as outlined in FIG.
40.
Although the invention has been described in detail for the
purposes of illustration, it is to be understood that such detail
is solely for that purpose and that variations can be made therein
by those skilled in the art without departing from the spirit and
scope of the invention except as it may be limited by the
claims.
__________________________________________________________________________
# SEQUENCE LISTING - (1) GENERAL INFORMATION: - (iii) NUMBER OF
SEQUENCES: 2 - (2) INFORMATION FOR SEQ ID NO:1: - (i) SEQUENCE
CHARACTERISTICS: #pairs (A) LENGTH: 20 base (B) TYPE: nucleic acid
(C) STRANDEDNESS: single (D) TOPOLOGY: linear - (ii) MOLECULE TYPE:
cDNA - (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1: # 20 GATA - (2)
INFORMATION FOR SEQ ID NO:2: - (i) SEQUENCE CHARACTERISTICS: #pairs
(A) LENGTH: 20 base (B) TYPE: nucleic acid (C) STRANDEDNESS: single
(D) TOPOLOGY: linear - (ii) MOLECULE TYPE: cDNA - (xi) SEQUENCE
DESCRIPTION: SEQ ID NO:2: # 20 CAGG
__________________________________________________________________________
* * * * *